VALUE IN HEALTH 15 (2012) 143–150

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/jval

The MacNew Heart Disease Health-Related Quality of Life

Questionnaire in Patients with Angina and Patients with Ischemic
Heart Failure
Stefan Höfer, PhD1,*, Atif Saleem, MD2, James Stone, MD, PhD3, Randal Thomas, MD4, Heather Tulloch, PhD5,
Neil Oldridge, PhD6
1
Department of Medical Psychology, Innsbruck Medical University, Innsbruck, Austria; 2Seton Hall and St. Joseph’s Regional Medical Center, Paterson, NJ,
USA; 3Faculty of Medicine, University of Calgary, Cardiac Wellness Institute, Calgary, AB, Canada; 4Mayo Clinic and Foundation, Rochester, MN, USA;
5
MINTO Prevention and Rehabilitation Centre, University of Ottawa Heart Institute, Ottawa, ON, Canada; 6University of Wisconsin School of Medicine and
Public Health and Aurora Cardiovascular Services, Aurora Sinai/Aurora St. Luke’s Medical Center, Milwaukee, WI, USA

A B S T R A C T

Objective: Patient-reported outcomes including health-related quality
of life are important in clinical care and research studies. The MacNew
Heart Disease health-related quality of life questionnaire has been val-
idated in English-speaking patients with myocardial infarction. The
aim of this study was to validate the MacNew in English-speaking pa-
tients with angina or ischemic heart failure. Methods: Canadian and
American patients with angina or ischemic heart failure completed the
MacNew, the Short Form-36 Health Survey, and the Hospital Anxiety
and Depression Scale. Results: We administered questionnaires to 276
patients with angina (mean age, 65.9 years) and 155 patients with isch-
emic heart failure (mean age, 70.3 years). The mean ⫾ SD MacNew
global score in patients with ischemic heart failure (5.1 ⫾ 1.2) was sta-
tistically (P ⬍ 0.001), but not clinically, poorer than in patients with
angina (5.3 ⫾ 1.1). The three-factor measurement model explained
46.1% of the observed variance in the MacNew in patients with angina
and 46.5% in patients with ischemic heart failure. Internal consistency
was ⱖ0.90, and test-retest reliability was ⱖ0.70 for each MacNew scale
and the a priori convergent and discriminative validity hypotheses
were confirmed in both diagnoses. The MacNew was highly accepted
by patients with little respondent or administrative burden.
Conclusions: The English version of the MacNew is reliable and valid
in patients with angina or ischemic heart failure. This permits health-
related quality of life outcome comparisons in patients with angina,
ischemic heart failure, and myocardial infarction with the MacNew
and provides a better understanding of the full range of health-related
quality of life outcomes.
Keywords: angina pectoris, coronary artery disease, health status, isch-
emic heart failure, patient-reported outcomes, quality of life, question-
naire.
Copyright © 2012, International Society for Pharmacoeconomics and
Outcomes Research (ISPOR). Published by Elsevier Inc.

which the instrument is validated and not with an “off-label diag-

Introduction
Patients with ischemic heart disease (IHD) present on a continuum
of events that includes the presence of risk factors, angina, myo-
cardial infarction (MI), and ischemic heart failure, often with
marked health-status deficits including poor health-related qual-
ity of life (HRQoL). As the patient’s own perspective of the impact
of disease and its treatment, patient-reported outcome measures
such as HRQoL have been recommended in both clinical care and
research studies by the National Heart, Lung and Blood Institute
[1], the U.S. Food and Drug Administration [2], and the European
Medicines Agency [3]. Generic HRQoL outcome measures permit
assessment of a wide range of aspects of life applicable to a variety
of health states and are useful in conducting general health survey
research [4]. Specific HRQoL outcome measures focus on disease-
relevant issues and are appropriate outcome measures in both
therapeutic intervention trials [4] and routine clinical care [5] but
should be used only in patients with the disease/diagnosis for
nosis.”
A core disease-specific HRQoL questionnaire approach with
adequate generalizability and sufficient specificity has been avail-
able for about two decades to make between-diagnosis outcome
comparisons, for example, in oncology [6,7] and rheumatology [8].
Between-diagnosis HRQoL outcome comparisons are not possible
with commonly used IHD diagnosis-specific questionnaires with
specific cues such as “your chest pain, chest tightness, or angina”
in the Seattle Angina Questionnaire [9] and “your heart failure” in
the Minnesota Living with Heart Failure Questionnaire [10]. In-
creasingly, patients with different IHD diagnoses receive similar
treatment, e.g., medication, percutaneous or surgical revascular-
ization, and referral to secondary prevention cardiac rehabilita-
tion, often with common therapeutic goals including symptom
management and improvement of HRQoL [1]. A core IHD-specific
HRQoL questionnaire with a cue such as “your heart problem”
would be useful in clinical practice and research studies allowing
between-diagnosis treatment outcome comparisons, would be ef-

* Address correspondence to: Stefan Höfer, Innsbruck Medical University, Speckbacherstr. 23/3, 6020 Innsbruck, Austria.
E-mail: stefan.hoefer@i-med.ac.at.
1098-3015/$36.00 – see front matter Copyright © 2012, International Society for Pharmacoeconomics and Outcomes Research (ISPOR).
Published by Elsevier Inc.
doi:10.1016/j.jval.2011.07.003
144 VALUE IN HEALTH 15 (2012) 143–150
ficient, and would provide a better understanding of the range of
HRQoL in patients with different IHD diagnoses.
The interviewer-administered Quality of Life after Myocardial
Infarction questionnaire was designed to evaluate how MI and its
treatment affects patients’ daily life and was validated [11] in Eng-
lish-speaking Canadian patients participating in a randomized
trial of cardiac rehabilitation [12]. This questionnaire was subse-
quently modified as the self-administered MacNew Heart Disease
HRQoL questionnaire and validated in English-speaking Austra-
lian patients with MI [13,14]. The MacNew has since been trans-
lated into 28 languages with independently published validation
studies in 8 of these languages in patients with MI (n ⬎ 4300),
angina (n ⬎ 1800), and ischemic heart failure (n ⬎ 550) [15–26].
As the psychometric properties demonstrated in patients with
MI [27] have been replicated in non–English-speaking patients
with angina or ischemic heart failure [17–20,22,23,26], the MacNew
may provide a core IHD HRQoL questionnaire for making between-
diagnosis comparisons of HRQoL outcomes. There are no valida-
tion studies, however, of the MacNew in English-speaking patients
with either angina or ischemic heart failure. The objective of this
study was therefore to investigate the reliability and validity of the
English version of the MacNew in patients with angina or ischemic
heart failure.
Methods
Patients
English-speaking patients with a diagnosis of angina or ischemic
heart failure were recruited in Canada and the United States as
part of the HeartQoL Project, an international survey of HRQoL in
more than 6300 patients with IHD in Australia, Europe, and North
America [28]. Institutional review boards approved the HeartQoL
Project at each site, and informed consent was obtained from all
subjects.
A convenience sample of patients 18 years of age and older able
to complete the self-administered battery of HRQoL instruments
in English, without serious psychiatric disorder, and not a sub-
stance abuser were eligible if they had a diagnosis of and being
treated for either:
Angina (Canadian Cardiovascular Society Grade II [slight limi-
tation of ordinary physical activity], grade III [marked limitation of
ordinary physical activity], or grade IV [inability to carry on any
physical activity without discomfort] [29]) with an objective mea-
sure of IHD (e.g., previous MI, exercise testing, echocardiography,
nuclear imaging, angiography) or ischemic heart failure (New York
Heart Association functional class II [patients with cardiac disease
resulting in slight limitation of physical activity], functional class
III [patients with cardiac disease resulting in marked limitation of
physical activity], or functional class IV [patients with cardiac dis-
ease resulting in the inability to perform any physical activity
without discomfort] [30]) with evidence of left ventricular dysfunc-
tion (ejection fraction ⱕ40% by invasive or noninvasive testing)
and an objective measure of IHD (e.g., previous MI, exercise test-
ing, echocardiography, nuclear imaging, angiography).
Questionnaires
The referring physician completed a questionnaire for routine di-
agnostic data, and all patients completed a self-report sociodemo-
graphic and clinical questionnaire, including health behaviors. In
addition, the Short Form-36 Health Survey (SF-36), the Hospital
Anxiety and Depression Scale (HADS), and the MacNew were com-
pleted by all patients at baseline and again 2 weeks later by the
first approximately 20% of the patients whom the site investigator
considered to be in a stable condition.
SF-36
The SF-36 is an internationally validated generic health survey
[31]. It consists of 36 items with 8 subscales summarized in a phys-
ical component summary (PCS) and a mental component sum-
mary (MCS) scale [32].
HADS
The HADS is a validated psychological screening instrument de-
signed to screen for symptoms of anxiety and depression [33]. It is
responsive and has been extensively used internationally in clin-
ical trials with scores of 8 or higher used to classify patients with
symptoms of depression or anxiety [34].
MacNew
The MacNew is designed to assess patient’s feelings about how
IHD affects daily functioning and contains 27 items with a global
HRQoL score and physical limitation, emotional, and social func-
tion subscales [13,14], with a summary of international results
available [35]. The items and scales are scored from 1 (low HRQoL)
to 7 (high HRQoL), and the minimal important difference (MID) on
each MacNew scale is 0.50 points [36]. The readability level of the
English MacNew has been established to be on average between
the fifth and sixth grade level [37].
Statistical analysis
Patient clinical and sociodemographic characteristics are de-
scribed as either dichotomous or continuous variables. HRQoL
scale scores (mean ⫾ SD) were calculated using established scor-
ing criteria for each instrument. Comparisons between the two
cardiac diagnostic groups were made using analysis of variance
(ANOVA) (continuous variables) and chi-square (␹2) (categorical
variables). Assumption for ANOVA (normality and homoschedac-
ity) was tested by Kolmorgorov-Smirnov, skewness, and kurtosis
statistics. If data did not meet the necessary criteria, the Welch
test (F) was used or the nonparametric Kruskal-Wallis H test was
applied.
Using recommended criteria [38] as described in detail below,
the conceptual model, reliability, validity, interpretability, and re-
spondent and administrative burden of the MacNew were as-
sessed for patients with angina and for patients with ischemic
heart failure. Face and content validity of the MacNew has been
previously documented [35].
The original factor structure identified by Valenti et al. [14] was
the basis for the confirmatory factor analysis that was carried out
using AMOS 18 [39]. Because ␹2 statistics are dependent on sample
size, a recommended range of parameters less sensitive to sample
size was selected. The ␹2/df, root mean square error of approximation
(RMSEA) and comparative fit index (CFI) were used to evaluate data
fit. Acceptable values are ⬍5 for ␹2/df [40], ⬎0.90 for the CFI and ⬍0.08
for the RMSEA, with an RMSEA of 0.0 indicating a perfect, 0.05 a good,
and 0.08 a reasonable fit [41]. Measurement errors of the items were
allowed to intercorrelate where appropriate. Further, differential
item functioning (DIF) was assessed [42]. A DIF analysis assesses
whether items are functioning equivalently across important cate-
gories, such as diagnosis (angina or ischemic heart failure). A mini-
mum difference in scores of 0.5 logits (P ⬍ 0.01) was used. DIF analysis
was performed using WINSTEPS 3.67 [43].
Internal consistency (Cronbach ␣) and 14-day test-retest reli-
ability (intraclass correlation coefficient [ICC]) were assessed using
r ⱖ 0.70 as the criterion value for both tests. With a priori predic-
tions hypothesizing strong correlations between similar SF-36 and
MacNew constructs and significantly lower correlations between
dissimilar constructs (⬍0.10 ⫽ absent, 0.10 – 0.29 ⫽ weak, 0.30 – 0.49
⫽ moderate, and ⱖ0.50 ⫽ strong), we tested the MacNew for con-
struct validity using Steiger’s test for comparing Pearson correla-
tions coefficients [44,45]. We assessed discriminative validity [46]
 VALUE IN HEALTH 15 (2012) 143–150 145

13.5% reported they were current smokers and 73.0% exercised less
Table 1 – Sociodemographic and clinical characteristics
than three times per week; 67.7% reported that their physician had
of patients with angina and patients with ischemic
told them they were hypertensive, 41.9% diabetic, and 65.2% with
heart failure.
high cholesterol. There were no patients with either Canadian Car-
Patient Angina IHF diovascular Society or New York Heart Association functional class I
characteristics (n ⫽ 276) (n ⫽ 155) per study design, and only a few with Canadian Cardiovascular So-
ciety angina class IV (4.0%) or New York Heart Association heart fail-
Age in years 65.9 ⫾ 11.4 70.3 ⫾ 12.3
ure functional class IV (3.9%).
Sex
Male 171 (62.0) 124 (80.0)
MacNew, SF-36 health status, anxiety, and depression
Female 105 (38.0) 31 (20.0)
(Table 2)
Family status
Single 45 (16.3) 20 (12.9) MacNew subscale scores could be calculated in all 431 patients,
Married 180 (65.2) 102 (65.8) SF-36 subscale scores in 409 patients (94.9%), and HADS scale
Other 41 (14.9) 32 (20.6) scores in 427 patients (99.1%).
Employment Patients with angina had better HRQoL than patients with isch-
White collar 122 (44.2) 75 (48.4) emic heart failure with significantly higher scores on the MacNew
Blue collar 96 (34.8) 45 (29.0) global, physical, and social subscales and on the SF-36 PCS but not
Education the MCS. The HADS anxiety scores and depression scores did not
Less than high 55 (19.9) 41 (26.5) differ significantly between patients with angina and those with
school ischemic heart failure.
High school 100 (36.2) 55 (35.5)
More than high 120 (43.5) 58 (37.4) MacNew questionnaire item characteristics (Table 3)
school
The missing MacNew item rate ranged from 0% (shortness of breath
Anxious (HADS 76 (27.5) 33 (21.3)
score ⱖ8) item) to 108 (39.1%) in patients with angina and 77 (49.7%) in patients
Depressed (HADS 56 (20.4) 36 (23.0) with ischemic heart failure (sexual activity item for both groups).
score ⱖ8) More than one MacNew item was missed by 42 of the patients (9.7%),
Body mass index 29.5 ⫾ 6.5 28.5 ⫾ 6.0 with only one patient missing as many as seven items. Therefore,
Smoker 36 (13.0) 21 (13.5) each MacNew subscale could be scored for every patient. There were
Functional status no floor effects for the MacNew global score or subscales in patients
CCS II: 154 (55.8) NYHA II: 56 (36.1) with either angina or ischemic heart failure. Ceiling effects were ob-
CCS III: 100 (36.2) NYHA III: 87 (56.1) served in ⱕ1.3% of the patients on the global scale and the physical
CCS IV: 11 (4.0) NYHA IV: 6 (3.9) and emotional subscales; ceiling effects were observed in 16 (5.8%) of
Hypertensive 193 (69.9) 105 (67.7) patients with angina and 6 (3.9%) in patients with ischemic heart
Diabetic 83 (30.1) 65 (41.9) failure on the social subscale for both groups.
Hypercholesterolemia 205 (74.3) 101 (65.2)
Physically inactive* 150 (54.5) 113 (73.0) MacNew factor structure (Table 4)
Values shown are mean ⫾ SD or number (%). Data missing when
Our findings confirm the multidimensionality of the MacNew and
⬍100%. CCS, Canadian Cardiovascular Society; HADS, Hospital
support the established three-factor MI model [13,14]. After allow-
Anxiety and Depression Scale; IHF, ischemic heart failure; NYHA,
ing measurement errors of the items to intercorrelate where ap-
New York Heart Association (functional class).
propriate, the three-factor model was supported for the combined
* Exercised less than three times/week.
data (␹2/df ⫽ 3.66; CFI ⫽ 0.938, RMSEA ⫽ 0.063, explained variance

testing the hypotheses that HRQoL would be 1) poorer in anxious

and depressed patients than those who were not using the HADS
Table 2 – MacNew, SF-36 physical component and
and 2) poorer in patients who perceived that their health had or
mental component scales, and HADS scores in patients
had not deteriorated over the past year using the SF-36 health
with angina and patients with ischemic heart failure.
transition item. We also report on the interpretability of the
MacNew scores relative to the published MacNew MID (ⱖ0.5 Angina IHF P value*
points) and on respondent and administrative burden.
MacNew
SPSS version 16.0 (SPSS, Inc., Chicago, Illinois) was used for all
Global 5.3 ⫾ 1.1 5.1 ⫾ 1.2 0.045
statistical analyses. Statistical significance was established at P ⱕ
Physical 5.3 ⫾ 1.2 4.9 ⫾ 1.4 0.003
0.05.
Emotional 5.3 ⫾ 1.1 5.2 ⫾ 1.2 0.434
Social 5.5 ⫾ 1.2 5.1 ⫾ 1.4 0.001
SF-36
Results PCS 39.7 ⫾ 11.3 35.8 ⫾ 10.1 0.001
MCS 49.5 ⫾ 10.8 49.1 ⫾ 10.9 0.715
Patient characteristics (Table 1) HADS
A convenience sample of 431 patients was recruited with a mean ⫾ Anxiety 5.8 ⫾ 3.5 5.4 ⫾ 3.4 0.336
SD age of 67.5 ⫾ 11.9 years, and 297 (68.9%) were male. The mean ⫾ Depression 4.6 ⫾ 3.5 5.1 ⫾ 3.8 0.303
SD age in the 276 patients with angina was 65.9 ⫾ 11.4 years and .
70.3 ⫾ 2.3 years in the 155 patients with ischemic heart failure. In the
HADS, Hospital Anxiety and Depression Scale; IHF, ischemic heart
angina group, 13.0% reported they were current smokers and 54.5%
failure; MCS, mental component summary; PCS, physical compo-
exercised less than three times per week; 69.9% reported that their
nent summary; SF-36, Short Form-36 Health Survey.
physician had told them they were hypertensive, 30.1% diabetic, and
* Kruskal-Wallis H test.
74.3% with high cholesterol. In the ischemic heart failure group,
146 VALUE IN HEALTH 15 (2012) 143–150

Table 3 – MacNew global and subscale floor and ceiling effects, internal consistency (Cronbach ␣), and test-retest
reliability in patients with angina and patients with ischemic heart failure.
MacNew

Global Physical Emotional Social

Angina
Floor effects, % 0 0 0 0
Ceiling effects, % 0.4 0.4 0.4 5.8
Cronbach ␣ 0.96 0.93 0.94 0.94
ICC (95% CI) (n ⫽ 62) 0.72 (0.57–0.82) 0.70 (0.55–0.81) 0.75 (0.62–0.84) 0.72 (0.57–0.82)
Ischemic heart failure
Floor effects, % 0 0 0 0
Ceiling effects, % 0.6 1.3 0.6 3.9
Cronbach ␣ 0.97 0.94 0.95 0.95
ICC (95% CI) (n ⫽ 22) 0.85 (0.68–0.94) 0.90 (0.68–0.96) 0.78 (0.55–0.90) 0.86 (0.69–0.94)

CI, confidence interval; ICC, intraclass correlation.

45.9%), for patients with angina (AP: ␹2/df ⫽2.41, CFI ⫽ 0.918,
RMSEA ⫽ 0.072, explained variance 46.1%), and for patients with
ischemic heart failure (IHF: ␹2/df ⫽ 2.08, CFI ⫽ 0.903, RMSEA ⫽
0.080, explained variance 46.5%). The intercorrelation of the fac-
tors ranged from 0.61 to 0.72. In both of the original MacNew factor
analyses based on patients with MI [13,14], a high proportion of
social subscale items (10 of 13) cross-loaded with more than one
subscale with factor loadings ⱖ0.40. In the current analysis, none
of the social subscale items cross-loaded in the physical or emo-
tional subscales with loadings ⱖ0.40 in patients with angina; only

Table 4 – Confirmatory factor analysis for each MacNew item in patients with angina and ischemic heart failure.
MacNew item Diagnosis

Angina Ischemic heart failure

Emo Phys Soc Emo Phys Soc

1. Frustrated 0.78 — — 0.81 — —

2. Worthless 0.64 — 0.26 0.79 — 0.13
3. Confident 0.60 — — 0.65 — —
4. Down in the dumps 0.85 — — 0.87 — —
5. Relaxed 0.78 — — 0.76 — —
6. Worn out 0.41 0.36 — 0.40 0.48 —
7. Happy with personal 0.77 — — 0.72 — —
life
8. Restless 0.69 — — 0.78 — —
9. Shortness of breath — 0.68 — — 0.83 —
10. Tearful 0.63 — — 0.65 — —
11. More dependent — — 0.65 — — 0.67
12. Social activities 0.05 0.35 0.30 0.32 0.32 0.11
13. Less confidence in you 0.29 — 0.49 0.44 — 0.38
14. Chest pain — 0.41 — — 0.53 —
15. Lack self-confidence 0.54 — 0.33 0.69 — 0.20
16. Aching legs — 0.40 — — 0.65 —
17. Sports/exercise limited — 0.99 — — 0.64 0.33
0.15
18. Frightened 0.72 — — 0.78 — —
19. Dizzy or lightheaded — 0.45 — — 0.43 —
20. Restricted or limited — 0.93 — — 0.67 0.34
0.07
21. Unsure about exercise — 0.68 0.08 — 0.46 0.42
22. Overprotective family — — 0.55 — — 0.56
23. Burden on others 0.20 — 0.68 0.42 — 0.46
24. Excluded — 0.47 0.38 — 0.30 0.61
25. Unable to socialize — 0.46 0.38 — 0.55 0.33
26. Physically restricted — 0.97 — — 0.66 0.36
0.07
27. Sexual activity — — 0.49 — — 0.55
Variance explained, % 19.5 18.3 8.3 23.2 14.1 9.2

Factor loading items shown in bold type differ from original factor loading in patients with myocardial infarction.
Emo, emotional subscale; Phys, physical subscale; Soc, social subscale.
 VALUE IN HEALTH 15 (2012) 143–150 147

Fig. 1 – Differential item function for angina (AP) and heart failure (HF) patients. Differential item functioning (DIF) measure
is presented in logits.

one of the social subscale items (item 21) cross-loaded in the phys- idation reports on the MacNew in other languages in patients with
ical subscale with a loading ⱖ0.40 in patients with ischemic heart angina and ischemic heart failure [24,26] as well as in patients
failure. Item 27 about sexual activity, which was not part of the with MI [13,14,16,21,26], providing support for the notion of the
original factor analysis [14], fitted best with the social subscale. No MacNew as a core IHD disease-specific HRQoL questionnaire.
DIF was observed for diagnosis, except for item 14 (DIF ⫽ ⫺0.55 The three-factor physical, emotional, and social subscale
logits, t(⫺6.21), P ⬍ 0.001, Fig. 1). structure of the original English MacNew in patients with MI was
confirmed in these patients with either angina or ischemic heart
MacNew reliability (Table 3) failure. In patients with angina, 46.1% of the total observed vari-
ance was explained, with 46.5% explained in patients with isch-
Internal consistency reliability of the MacNew global and each
emic heart failure. The results of this confirmatory factor analysis
subscale was confirmed with all Cronbach ␣ values ⬎0.90 in pa-
indicate that the emotional and physical subscales are most con-
tients with each diagnosis. As a measure of test-retest reliability,
sistent with the original factor structure identified in patients with
the ICC was always ⱖ0.70, ranging from 0.70 (physical subscale) to
MI [14]; only two emotional items (items 13 and 23) and one phys-
0.75 (emotional subscale) in patients with angina and from 0.78
ical item (item 6) in patients with angina and only one physical
(emotional subscale) to 0.90 (physical subscale) in patients with
subscale item (item 24) in patients with ischemic heart failure
ischemic heart failure.
differed from the original factor analysis. Cross-loadings of ⱖ0.40
were allowed in the original model with 11 of the 13 social subscale
MacNew convergent validity (Table 5) items associated with either physical or emotional items [14],
The correlations between similar MacNew and SF-36 scales none of which were substantiated in the current study in patients
(MacNew physical and PCS; MacNew emotional and MCS) were with either angina or ischemic heart failure. In fact, the item about
significant (P ⬍ 0.001), strong, and in the hypothesized direction. social activities (item 12) showed cross-factor loadings of ⬍0.40,
The correlations between dissimilar MacNew and SF-36 scales which is quite different from the original model in which it was the
were significantly lower than between the similar scales (all one- only item with cross-loadings of ⱖ0.40 on all three subscales [14].
sided P values ⬍0.05). Taken together, these results confirm the
convergent validity of the MacNew scales in patients with angina
and ischemic heart failure.
Table 5 – Convergent validity of the MacNew physical
MacNew discriminative validity (Table 6) and emotional subscales with the SF-36 physical
component and mental component scales in patients
Discriminative validity of the MacNew was confirmed for each
with angina (n = 262) and patients with ischemic heart
criterion (P ⬍ 0.001) in patients with angina and in patients with
failure (n = 147).
ischemic heart failure. In addition, all MacNew score differences
exceeded the MID criterion of ⱖ0.5 between 1) the two diagnostic MacNew MacNew One-sided
groups, 2) those improving or staying the same and those deteri- physical emotional P value*
orating on the SF-36 health transition item (always ⬎0.88 in pa- subscale subscale
tients with angina and ⬎1.24 in patients with ischemic heart fail-
ure), 3) those with and without anxiety (always ⬎1.02 in patients Angina
with angina and ⬎1.27 in patients with ischemic heart failure), and SF-36 PCS 0.728† 0.510† ⬍0.001
4) those with and without depression (⬎1.45 in patients with an- SF-36 MCS 0.490† 0.731† ⬍0.001
gina and ⬎1.91 in patients with ischemic heart failure). One-sided P value* ⬍0.001 ⬍0.001
Ischemic heart failure
SF-36 PCS 0.743† 0.632† 0.035
SF-36 MCS 0.584† 0.750† 0.004
Discussion One-sided P value* 0.007 0.026
The English-language MacNew Heart Disease HRQoL question- Strong correlations, r ⱖ 0.50.
naire meets the recommended psychometric criteria for the con- SF-36 MCS, Short Form-36 Health Survey mental component sum-
ceptual model, reliability, validity, interpretability, and burden mary; SF-36 PCS, Short Form-36 Health Survey physical component
[38] in patients with either angina or ischemic heart failure. The summary.
expected patterns in MacNew global and each subscale scores * Steiger’s test for comparing Pearson correlation coefficients.
were confirmed as patients with ischemic heart failure had poorer †
Pearson correlation coefficients; P values ⬍0.001.
scores than patients with angina. These findings substantiate val-
148 VALUE IN HEALTH 15 (2012) 143–150

Table 6 – Discriminant validity of MacNew global score and subscale scores by SF-36 health status transition and HADS
anxiety and depression in patients with angina and patients with ischemic heart failure.
MacNew

Global Physical Emotional Social

Angina
SF-36 health transition
Improve 5.62 ⫾ 0.9* 5.67 ⫾ 1.0* 5.55 ⫾ 1.0* 5.84 ⫾ 1.0*
No change 5.61 ⫾ 0.8† 5.53 ⫾ 0.9† 5.56 ⫾ 0.8† 5.89 ⫾ 1.0†
Deteriorate 4.65 ⫾ 1.2 4.51 ⫾ 1.3 4.66 ⫾ 1.3 4.73 ⫾ 1.3
Anxiety
No 5.63 ⫾ 0.9 5.53 ⫾ 1.1 5.65 ⫾ 0.9 5.80 ⫾ 1.0
Yes 4.49 ⫾ 1.1‡ 4.58 ⫾ 1.3‡ 4.32 ⫾ 1.1‡ 4.77 ⫾ 1.4‡
Depression
No 5.63 ⫾ 0.8 5.56 ⫾ 1.0 5.63 ⫾ 0.8 5.83 ⫾ 0.9
Yes 4.11 ⫾ 1.0‡ 4.10 ⫾ 1.2‡ 3.92 ⫾ 1.0‡ 4.24 ⫾ 1.3‡
Ischemic heart failure
SF-36 health transition
Improve 5.51 ⫾ 0.9* 5.41 ⫾ 1.2* 5.57 ⫾ 1.0* 5.51 ⫾ 1.1*
No change 5.51 ⫾ 0.9† 5.29 ⫾ 1.2† 5.60 ⫾ 1.0† 5.60 ⫾ 1.2†
Deteriorate 4.20 ⫾ 1.1 3.88 ⫾ 1.2 4.33 ⫾ 1.2 4.05 ⫾ 1.3
Anxiety
No 5.41 ⫾ 1.0 5.16 ⫾ 1.3 5.56 ⫾ 0.9 5.38 ⫾ 1.2
Yes 3.95 ⫾ 1.2‡ 3.88 ⫾ 1.3‡ 3.82 ⫾ 1.2‡ 3.96 ⫾ 1.5‡
Depression
No 5.53 ⫾ 0.9 5.36 ⫾ 1.0 5.62 ⫾ 0.8 5.56 ⫾ 1.1
Yes 3.61 ⫾ 0.8‡ 3.27 ⫾ 0.8‡ 3.70 ⫾ 1.0‡ 3.44 ⫾ 0.9‡

Values shown are mean ⫾ SD. Analysis of variance (Welch test) or Kruskal-Wallis H test.
HADS, Hospital Anxiety and Depression Scale; SF-36, Short Form-36 Health Survey.
* Improve vs. deteriorate; P ⬍ 0.05.
†
No change vs. deteriorate; P ⬍ 0.05.
‡
P ⬍ 0.001 in each diagnosis.

Item 27 about sexual activity was not part of the original factor
structure [13,14] and, consistent with more recent international
factor analysis studies [17, 21,23], fell into the social subscale
rather than the physical subscale. A potential explanation for
these findings is that the MacNew social items in general relate to
social situations, either in a physical or interpersonal (emotional)
context. These observations and concerns expressed previously
about the factor structure of the MacNew [47,48] suggest that re-
visiting the MacNew from both a modeling perspective allocating
items only to their main subscale and an item reduction perspec-
tive would be productive. This is in the process of being addressed.
No differential item function was detected except for item 14. Item
14 addresses “how often during the past two weeks have you ex-
perienced chest pain while doing your day-to-day activities,”
which was more likely to be endorsed by angina patients than
ischemic heart failure patients.
Consistent with MacNew internal consistency reports in other
languages [17,19,21–27], the Cronbach ␣ was ⬎0.90 on all scales in
each diagnosis allowing for both group and individual HRQoL com-
parisons [38,49]. When administered twice, 14 days apart, test-
retest reliability was confirmed in both diagnoses with an ICC
⬎0.70 on all MacNew scales, also consistent with MacNew reports
in other languages [22–24,26]. The lower test-retest reliabilities ob-
served in patients with angina than in those with ischemic heart
failure are most likely explained by the fluctuating nature of chest
pain, even when well managed.
Convergent and discriminative validity of the MacNew was
confirmed in patients with angina or ischemic heart failure. Sim-
ilar MacNew and SF-36 scales (MacNew physical and PCS; MacNew
emotional and MCS) had significantly high correlations. Dissimilar
scales (MacNew physical and MCS; MacNew emotional and PCS)
had significantly lower correlations than those of the similar
scales. The MacNew discriminated significantly by SF-36 health
transition status, anxiety, and depression in both diagnostic
groups. Patients with perceived health deterioration over the past
year had significantly poorer MacNew global HRQoL and subscale
scores than patients whose health had improved or stayed the
same, with the difference exceeding the MacNew MID criterion of
0.5 points [36]. Similar MacNew observations were observed by the
presence or absence of anxiety or depression, with the differences
also exceeding the MID criterion in each instance. As an extension
of these observations, the sample size required to demonstrate
HRQoL differences between depressed and nondepressed patients
with the specific MacNew HRQoL instrument would be only 40% of
that estimated for the generic SF-36 HRQoL instrument (MacNew,
n ⫽ 8 per group vs. SF-36, n ⫽ 22 per group), and this observation
holds true for depression and for health transition status. This is
important in terms of the potential usefulness of the MacNew
because sample size is a major consideration when designing
high-cost randomized trials.
Item relevance and minimal respondent and administrative
burden are fundamental to the acceptance, implementation, and
utilization of HRQoL instruments by patients, clinicians, and re-
searchers [38]. That only 1 of the 431 patients missed no more than
7 of the 27 MacNew items in this study clearly demonstrates the
relevance of items to patients, and MacNew global and subscale
scores could be calculated for each of the 431 patients. Patients
with cardiovascular disease frequently report fatigue [50]. “Feeling
worn out” (MacNew item 6) was reported to be the number one
problem by patients in both diagnostic groups in the current study,
suggesting that fatigue can be considered a potential treatment
focus for many patients with angina or ischemic heart failure.
Finally, the self-administered MacNew is typically completed in
 VALUE IN HEALTH 15 (2012) 143–150
⬍10 minutes and is simple to score, resulting in minimal respon-
dent and administrative burden [35].
The major limitation to this study is that the parent HeartQoL
Project was designed as a cross-sectional survey for recruiting
convenience samples of patients at each site [28]. As a result, we
are unable to generate and report responsiveness statistics for the
MacNew in these patients with either angina or ischemic heart
failure. Although some ceiling effect was observed on the social
subscale, the minimal floor and ceiling effects on the subscales
should permit measurement of change, both deterioration and
improvement, in HRQoL. The MacNew consistently has been
shown to be responsive after different treatment options in pa-
tients with angina, myocardial infarction, or ischemic heart failure
such as cardiac rehabilitation [24,51–56], revascularization [57,58],
and medication [59], and we see no reason why the MacNew would
not be responsive in other patients with angina or ischemic heart
failure.
The results of this report in patients either with angina or isch-
emic heart failure demonstrate that the English version of the
MacNew meets the psychometric standards for measurement, re-
liability, validity, interpretability, item relevance, and burden recom-
mended for HRQoL instruments [38], and, although responsiveness
could not be evaluated, this has been repeatedly demonstrated in
other studies in patients with either angina or ischemic heart failure.
This study extends previous international validation studies in other
languages in patients with angina, MI, and ischemic heart failure
[24,26], substantiating the suggesting of the MacNew as a potential
core IHD-specific HRQoL instrument. A core IHD-specific HRQoL
questionnaire will provide a better understanding of the range in
HRQoL in patients with IHD that would potentially optimize clinical
service provision and clinical medicine and social science out-
come research efforts when assessing HRQoL or making HRQoL
outcome comparisons across patients with angina, MI, or isch-
emic heart failure.
Acknowledgments
We thank T. Loessin, J. Miller, C. Nelson, and J. Paddock, who re-
cruited patients and entered data. We also thank all the patients
who provided extensive personal information. Source of financial
support: The HeartQoL Project received partial funding from the
European Association of Cardiovascular Prevention and Rehabili-
tation and the European Society of Cardiology.
REFERENCES
[1] Krumholz HM, Peterson ED, Ayanian JZ, et al. Report of the National
Heart, Lung, and Blood Institute working group on outcomes research
in cardiovascular disease. Circulation 2005;111:3158 – 66.
[2] U.S. Department of Health and Human Services, Food and Drug
Administration. Guidance for industry. Patient-reported outcome
measures: Use in medical product development to support labeling
claims. Washington, DC. 2009. Available from: http://www.fda.gov/
downloads/Drugs/GuidanceComplianceRegulatoryInformation/
Guidances/UCM193282.pdf. [January, 21, 2010].
[3] Committee for Medicinal Products for Human Use. Reflection paper on
the regulatory guidance for the use of health-related quality of life
(HRQL) measures in the evaluation of medicinal products. European
Medicines Agency; 2005. Available from: http://www.emea.europa.eu/
pdfs/human/ewp/13939104en.pdf. [Accessed January, 21, 2010].
[4] Testa MA, Simonson DC. Assessment of quality of life outcomes.
N Engl J Med 1996;334:835– 40.
[5] Velikova G, Booth L, Smith AB, et al. Measuring quality of life in
routine oncology practice improves communication and patient well-
being: a randomized controlled trial. J Clin Oncol 2004;22:714 –24.
[6] Aaronson NK, Ahmedzai S, Bergman B, et al. The European
Organization for Research and Treatment of Cancer QLQ-C30: a
quality-of-life instrument for use in international clinical trials in
oncology. J Natl Cancer Inst 1993;85:365–76.
149
[7] Cella DF, Tulsky DS, Gray G, et al. The Functional Assessment of
Cancer Therapy scale: development and validation of the general
measure. J Clin Oncol 1993;11:570 –9.
[8] Bellamy N, Buchanan W, Goldsmith C, et al. Validation study of
WOMAC: a health status instrument for measuring clinically
important patient relevant outcomes to anti-rheumatic drug therapy
in patients with osteoarthritis of the hip or knee. J Rheumatology 1988;
15:1833– 40.
[9] Spertus JA, Winder JA, Dewhurst TA, et al. Development and
evaluation of the Seattle Angina Questionnaire: a new functional
status measure for coronary artery disease. J Am Coll Cardiol 1995;25:
333– 41.
[10] Rector TS, Kubo SH, Cohn JN. Patients’ self-assessment of their
congestive heart failure: Part 2. Content, reliability, and validity of a
new measure, the Minnesota Living with Heart Failure questionnaire.
Heart Failure 1987;3:198 –209.
[11] Hillers TK, Guyatt GH, Oldridge N, et al. Quality of life after myocardial
infarction. J Clin Epidemiol 1994;47:1287–96.
[12] Oldridge N, Guyatt G, Jones N, et al. Effects on quality of life with
comprehensive rehabilitation after acute myocardial infarction. Am J
Cardiol 1991;67:1084 –9.
[13] Lim LL-Y, Valenti LA, Knapp JC, et al. A self-administered quality of
life questionnaire after acute myocardial infarction. J Clin Epidemiol
1993;46:1249 –56.
[14] Valenti L, Lim L, Heller RF, Knapp J. An improved questionnaire for
assessing quality of life after myocardial infarction. Qual Life Res 1996;
5:151– 61.
[15] Asadi-Lari M, Javadi H, Melville M, et al. Adaptation and
administration of the MacNew quality of life questionnaire after
myocardial infarction in an Iranian population. Health and Qual Life
2003;1:23.
[16] Brotons Cuixart C, Ribera Sole A, Permanyer Miralda G, et al.
Adaptation of the MacNew QLMI quality of life questionnaire after
myocardial infarction to be used in the Spanish population [in
Spanish]. Med Clin (Barc) 2000;115:768 –71.
[17] Daskapan A, Höfer S, Oldridge N, et al. The validity and reliability of
the Turkish version of the MacNew Heart Disease Questionnaire in
patients with angina. J Eval Clin Pract 2008;14:209 –13.
[18] De Gucht V, Van Elderen T, Van Der Kamp L, Oldridge N. Quality of life
after myocardial infarction: translation and validation of the MacNew
questionnaire for a Dutch population. Qual Life Res 2004;13:1483– 8.
[19] Hiller A, Helvik AS, Kaasa S, Slordahl SA. Psychometric properties of
the Norwegian MacNew Heart Disease health-related quality of life
inventory. Eur J Cardiovasc Nurs 2010;9:146 –52.
[20] Höfer S, Benzer W, Schussler G, et al. Health-related quality of life in
patients with coronary artery disease treated for angina: validity and
reliability of German translations of two specific questionnaires. Qual
Life Res 2003;12:199 –212.
[21] Höfer S, Benzer W, Brandt D, et al. MacNew Heart Disease
questionnaire after myocardial infarction: the German version. Z Klin
Psychol Psychother 2004;33:270 – 80.
[22] Höfer S, Schmid JP, Frick M, et al. Psychometric properties of the
MacNew heart disease health-related quality of life instrument in
patients with heart failure. J Eval Clin Pract 2008;14:500 – 6.
[23] Leal A, Paiva C, Höfer S, et al. Evaluative and discriminative properties
of the Portuguese MacNew health-related quality of life questionnaire.
Qual Life Res 2005;14:2335– 41.
[24] Maes S, De Gucht V, Goud R, et al. Is the MacNew quality of life
questionnaire a useful diagnostic and evaluation instrument for
cardiac rehabilitation? Eur J Cardiovasc Prev Rehabil 2008;15:516 –20.
[25] Nakajima KM, Rodrigues RC, Gallani MC, et al. Psychometric
properties of MacNew Heart Disease Health-related Quality of Life
Questionnaire: Brazilian version. J Adv Nurs 2009;65:1084 –9.
[26] Yu DS, Thompson DR, Yu CM, Oldridge NB. Validation of the Chinese
version of the MacNew heart disease health-related quality of life
questionnaire. J Eval Clin Pract 2008;14:326 –35.
[27] Höfer S, Anelli-Monti M, Berger T, et al. Psychometric properties of an
established heart disease specific health-related quality of life
questionnaire for pacemaker patients. Qual Life Res 2005;14:1937– 42.
[28] Oldridge N, Saner H, McGee HM. The Euro Cardio-QoL Project. An
international study to develop a core heart disease health-related
quality of life questionnaire, the HeartQoL. Eur J Cardiovasc Prev
Rehabil 2005;12:87–94.
[29] Campeau L. Grading of angina pectoris. Circulation 1976;54:522–3.
[30] Criteria Committee of the New York Heart Association. Nomenclature
and Criteria for Diagnosis of Diseases of the Heart and Great Vessels.
9th ed. Dolgin M, editor. Boston: Little, Brown and Company, 1994.
[31] Ware JE Jr. SF-36 health survey update. Spine 2000;25:3130 –9.
[32] Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey
(SF-36). I. Conceptual framework and item selection. Med Care 1992;
30:473– 83.
[33] Zigmond AS, Snaith RP. The hospital anxiety and depression scale.
Acta Psychiatr Scand 1983;67:361–70.
150 VALUE IN HEALTH 15 (2012) 143–150
[34] Bjelland I, Dahl AA, Haug TT, Neckelmann D. The validity of the
Hospital Anxiety and Depression Scale. An updated literature review.
J Psychosom Res 2002;52:69 –77.
[35] Höfer S, Lim LL, Guyatt GH, Oldridge N. The MacNew Heart Disease
Health-related Quality of Life Instrument: a summary. Health Qual
Life Outcomes 2004;2:3.
[36] Dixon T, Lim L, Oldridge N. The MacNew health-related quality of life
instrument: reference data for users. Qual Life Res 2002;11:173– 83.
[37] Gromoske A, Oldridge N, Brondino MJ. Readability of three heart
disease health-related quality of life questionnaires. J Cardiopulm
Rehabil Prevent 2011;31:245– 8.
[38] Scientific Advisory Committee of Medical Outcomes Trust. Assessing
health status and quality-of-life instruments: attributes and review
criteria. Qual Life Res 2002;11:193–205.
[39] Byrne B. Structural Equation Modeling With AMOS: Basic Concepts,
Applications, and Programming. Mahwah, NJ: Lawrence Erlbaum
Associates, 2001.
[40] Schumacker RE, Lomax RG. A beginner’s guide to structural equation
modeling. Mahwah, NJ: Lawrence Erlbaum Associates, 2004.
[41] Browne M, Cudeck R. Alternative ways of assessing model fit. In:
Bollen K, Long J, eds. Testing Structural Equation Models. Newbury
Park, CA: Sage, 1993.
[42] Holland PW, Wainer H. Differential Item Functioning. Hillsdale, NJ:
Lawrence Erlbaum Associates, 1993.
[43] Linacre JM. A User’s Guide to Winsteps/Ministeps Rasch Model
Computer Programs 2008.
[44] Steiger J. Tests for comparing elements of a correlation matrix.
Psychol Bull 1980;87:245–51.
[45] Cohen J. Statistical Power Analysis for the Behavioral Sciences.
Hillsdale, NJ: Lawrence Erlbaum Associates, 1988.
[46] Hays RD, Anderson RT, Revicki D. Assessing reliability and validity of
measurement in clinical trials. In: Staquet MJ, Hays RD, Fayers P,
editors. Quality of Life Assessment in Clinical Trials: Methods and
Practice. New York: Oxford University Press, 1998.
[47] Dempster M, Donnelly M, O’Loughlin C. The validity of the MacNew
Quality of Life in heart disease questionnaire. Health Qual Life
Outcomes 2004;2:6.
[48] Ribera A, Permanyer-Miralda G, Alonso J, et al. Is psychometric scoring
of the MacNew Quality of Life after Myocardial Infarction
questionnaire superior to the clinimetric scoring? A comparison of the
two approaches. Qual Life Res 2006;15:357– 65.
[49] Bland JM, Altman DG. Cronbach’s alpha. BMJ 1997;314:572.
[50] Casillas JM, Damak S, Chauvet-Gelinier JC, et al. Fatigue in patients with
cardiovascular disease. Ann Readapt Med Phys 2006;49:309 –19, 92– 402.
[51] Benzer W, Platter M, Oldridge NB, et al. Short-term patient-reported
outcomes after different exercise-based cardiac rehabilitation
programmes. Eur J Cardiovasc Prev Rehabil 2007;14:441–7.
[52] Dalal HM, Evans PH, Campbell JL, et al. Home-based versus
hospital-based rehabilitation after myocardial infarction: a
randomized trial with preference arms--Cornwall Heart Attack
Rehabilitation Management Study (CHARMS). Int J Cardiol 2007;119:
202–11.
[53] Höfer S, Kullich W, Graninger U, et al. Cardiac rehabilitation in
Austria: Long term health-related quality of life outcomes. Health
Qual Life Outcomes 2009;7:99.
[54] Reid R, Dafoe W, Morrin L, et al. Impact of program duration and
contact frequency on efficacy and cost of cardiac rehabilitation: a
randomized trial. Am Heart J 2005;149:862– 8.
[55] Wisloff U, Stoylen A, Loennechen JP, et al. Superior cardiovascular
effect of aerobic interval training versus moderate continuous training
in heart failure patients: a randomized study. Circulation
2007;115:3086 –94.
[56] Eder B, Hofmann P, von Duvillard SP, et al. Early 4-week cardiac
rehabilitation exercise training in elderly patients after heart surgery.
J Cardiopulm Rehabil Prev 2010;30;85–92.
[57] Benzer W, Höfer S, Oldridge NB. Health-related quality of life in
patients with coronary artery disease after different treatments for
angina in routine clinical practice. Herz 2003:28:421– 8.
[58] Höfer S, Kullich W, Graninger U, et al. Cardiac rehabilitation in
Austria: short term quality of life improvements in patients with heart
disease. Wien Klin Wochenschr 2006;118:744 –53.
[59] Marazzi G, Gebara O, Vitale C, et al. Effect of trimetazidine on quality
of life in elderly patients with ischemic dilated cardiomyopathy. Adv
Ther 2009;26:455– 61.